摘要
目标检测是计算机视觉研究中的热门问题,其中加速区域卷积神经网络(Faster R-CNN)对目标检测具有指导意义。针对Faster R-CNN算法在目标检测中准确率不高的问题,先对数据进行增强处理;然后对提取的特征图进行裁剪,利用双线性插值法代替感兴趣区域池化操作,分类时采用软非极大值抑制(Soft-NMS)算法。实验结果表明,该算法在PASCAL VOC2007、PASCAL VOC07+12数据集下的准确率分别为76.40%和81.20%,相较Faster R-CNN算法分别提升了6.50个百分点和8.00个百分点。没有进行数据增强的情况下,在COCO 2014数据集上的准确率相较Faster R-CNN算法提升了2.40个百分点。
Object detection is a hot topic in computer vision research,among which faster region-based convolutional neural network(Faster R-CNN)has guiding significance for object detection.Aiming at the problem of the low accuracy of the Faster R-CNN algorithm in object detection,the data is enhanced first.Then,the extracted feature map is trimmed,and bilinear interpolation is used to replace the region of interest pooling operation.Soft-nonmaximum suppression(Soft-NMS)algorithm is used for classification.Experimental results show that the accuracy of the algorithm is 76.40%and 81.20%in PASCAL VOC2007 and PASCAL VOC07+12 datasets,which is 6.50 percentage points and 8.00 percentage points higher than that of the Fast R-CNN algorithm,respectively.Without data enhancement,the accuracy on the COCO 2014 dataset is improved by 2.40 percentage points compared with that of the Faster R-CNN algorithm.
作者
周兵
李润鑫
尚振宏
李晓武
Zhou Bing;Li Runxin;Shang Zhenhong;Li Xiaowu(Faculty of Information Engineering and Automation,Kunming Unirersity of Science and Technology,Kumming,Yumnan 650500,China.)
出处
《激光与光电子学进展》
CSCD
北大核心
2020年第10期97-104,共8页
Laser & Optoelectronics Progress
基金
国家自然科学基金(61462052)
昆明理工大学引进人才科研启动项目(KKSY201603016)。
